AI4EOSC - AI for the European Open Science Cloud


The AI4EOSC (Artificial Intelligence for the European Open Science Cloud) project aims to deliver an enhanced set of services for the development of Artificial Intelligence (AI), Machine Learning (ML) and Deep Learning (DL) models and applications. The project builds on the DEEP-Hybrid-DataCloud and EOSC computing services to provide a specialized platform, offering customizable components for tailor made deployments while adapting to evolving user needs. AI4EOSC thereby aims to increase of the number of advanced, high level, customizable services available through the EOSC portal, thus serving as a catalyst for researchers, facilitating collaboration, simplifying access to high-end pan-European resources and reducing computing time. These goals are to be reached by increasing the service offer in the EU landscape by expanding the existing European Open Science Cloud (EOSC) ecosystem. The services will allow for advanced features such as distributed, federated and split learning; provenance metadata; event-driven data processing services or provisioning of services based on serverless computing.

The project will focus on tools to provide AI, ML and DL services by integrating real life use cases to aid in the design process and showcase the afore-mentioned functionalities. AI4EOSC bases its activities on the technological framework delivered by the DEEP-Hybrid-DataCloud H2020 project. The DEEP platform (provided through the EOSC portal 2) is a production-ready system that is being effectively used by researchers in the EU to train and develop machine learning and deep learning models.

Throughout project development, a special emphasis will be placed on ensuring that all the research outputs and sub-products (data, models, metadata, publications, etc.) adhere to the FAIR data and research principles.

Project Partners

The AI4EOSC consortium has been assembled to ensure a balanced and complementary set of partners with backgrounds in research, development, technology and innovation. The consortium comprises several of the most active institutions in terms of development, implementation, deployment and operation of distributed pan-European e-infrastructures as well as experienced and highly innovative SMEs with a untapped potential in the field of AI. All partners involved in the project are highly experienced in software development, several of them having previously developed key components in EU e-Infrastructure.

The consortium consists of 10 partners from academia (including the project coordinator CSIC, KIT, IISAS, UPV, LIP, INFN and PSNC) and industry (Predictia, MicroStep-MIS and WODR).

IIP Contribution

The IIP joins this endeavour by providing a use case on automated thermography consisting of thermal images of city infrastructure such as buildings and the ground above district heating networks. These will form a basis to test the platform’s functionality and proficiency in incorporating new AI-based models to - in this case - detect thermal bridges on buildings and common thermal anomalies.

Associated Publications

How Does Feature Engineering Impact UAV-based Multispectral Semantic Segmentation? An RGB and Thermal Image Ablation Study
Vollmer, E.; Benz, M.; Kahn, J.; Klug, L.; Volk, R.; Schultmann, F.; Götz, M.
2024, June 12. Helmholtz Artificial Intelligence Conference (Helmholtz AI 2024), Düsseldorf, Germany, June 12–14, 2024
AI4EOSC - D6.2 Intermediate status report about integration of pilot applications
Kozlov, V.; Sainz-Pardo, J.; Berberi, L.; Alibabaei, K.; Vollmer, E.; Błaszczak, M.; Krzyżanek, M.; Smok, J.; Bartok, J.; Sisan, P.; Papadopoulos, G.; Izquierdo, P.
2024. Zenodo. doi:10.5281/zenodo.10729327
Federated Learning for Urban Energy Efficiency: Detecting Thermal with UAV-based Imaging and AI
Duda, L. J.; Alibabaei, K.; Vollmer, E.; Klug, L.; Benz, M.; Kozlov, V.; Volk, R.; Goetz, M.; Schultmann, F.; Streit, A.
2024, June 12. Helmholtz Artificial Intelligence Conference (Helmholtz AI 2024), Düsseldorf, Germany, June 12–14, 2024
AI in multispectral image analysis: Implementing a deep learning model for the segmentation of common thermal urban features to assist in the automation of infrastructure-related maintenance
Vollmer, E.; Klug, L.; Volk, R.; Schultmann, F.
2024, March 21. 4th Artificial Intelligence in Architecture, Engineering and Construction Conference (2024), Helsinki, Finland, March 20–21, 2024
D6.1 : Analysis of user applications, collection of requirements
Berberi, L.; Heredia, I.; Kozlov, V.; Vollmer, E.; Sainz-Pardo, J.; Bartok, J.; Fojud, A.; Blaszczak, M.; Rausch, A.; Alibabaei, K.
2023. Zenodo. doi:10.5281/zenodo.7635453
UAV-based Thermography: Using AI with Multispectral Data
Vollmer, E.
2023, December 7. ANERIS Workshops on AI Basics for Image Processing (2023), Online, November 28–December 7, 2023